US12003254B2 - Method and apparatus for transmitting data on basis of polar coding in wireless communication system - Google Patents
Method and apparatus for transmitting data on basis of polar coding in wireless communication system Download PDFInfo
- Publication number
- US12003254B2 US12003254B2 US17/297,705 US201917297705A US12003254B2 US 12003254 B2 US12003254 B2 US 12003254B2 US 201917297705 A US201917297705 A US 201917297705A US 12003254 B2 US12003254 B2 US 12003254B2
- Authority
- US
- United States
- Prior art keywords
- channel
- information blocks
- action
- learning
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active, expires
Links
- 238000000034 method Methods 0.000 title claims abstract description 198
- 238000004891 communication Methods 0.000 title claims abstract description 82
- 230000009471 action Effects 0.000 claims abstract description 164
- 101000741965 Homo sapiens Inactive tyrosine-protein kinase PRAG1 Proteins 0.000 claims abstract description 27
- 102100038659 Inactive tyrosine-protein kinase PRAG1 Human genes 0.000 claims abstract description 27
- 125000004122 cyclic group Chemical group 0.000 claims abstract description 9
- 230000015654 memory Effects 0.000 claims description 57
- 230000005540 biological transmission Effects 0.000 claims description 54
- 238000010801 machine learning Methods 0.000 claims description 22
- 230000003247 decreasing effect Effects 0.000 claims description 9
- 238000010586 diagram Methods 0.000 description 54
- 239000010410 layer Substances 0.000 description 45
- 241000169170 Boreogadus saida Species 0.000 description 43
- 238000013528 artificial neural network Methods 0.000 description 42
- 230000009897 systematic effect Effects 0.000 description 34
- 230000008859 change Effects 0.000 description 26
- 230000006870 function Effects 0.000 description 25
- 230000007423 decrease Effects 0.000 description 24
- 238000005259 measurement Methods 0.000 description 24
- 230000008569 process Effects 0.000 description 18
- 238000005516 engineering process Methods 0.000 description 16
- 239000013598 vector Substances 0.000 description 16
- 230000002787 reinforcement Effects 0.000 description 14
- 239000011159 matrix material Substances 0.000 description 11
- 238000005457 optimization Methods 0.000 description 11
- 238000012545 processing Methods 0.000 description 10
- 206010009944 Colon cancer Diseases 0.000 description 8
- 239000003795 chemical substances by application Substances 0.000 description 8
- 238000012937 correction Methods 0.000 description 8
- 230000002829 reductive effect Effects 0.000 description 7
- 230000000306 recurrent effect Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000013473 artificial intelligence Methods 0.000 description 4
- 230000001186 cumulative effect Effects 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 238000010295 mobile communication Methods 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 230000007704 transition Effects 0.000 description 4
- 238000004904 shortening Methods 0.000 description 3
- 230000008054 signal transmission Effects 0.000 description 3
- 238000012546 transfer Methods 0.000 description 3
- 241000700159 Rattus Species 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 210000004027 cell Anatomy 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000005562 fading Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000003780 insertion Methods 0.000 description 2
- 230000037431 insertion Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000011017 operating method Methods 0.000 description 2
- 230000010287 polarization Effects 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 102100022734 Acyl carrier protein, mitochondrial Human genes 0.000 description 1
- 241000760358 Enodes Species 0.000 description 1
- 101000678845 Homo sapiens Acyl carrier protein, mitochondrial Proteins 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000002346 layers by function Substances 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 230000001151 other effect Effects 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004984 smart glass Substances 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000005406 washing Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M13/00—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
- H03M13/03—Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words
- H03M13/05—Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words using block codes, i.e. a predetermined number of check bits joined to a predetermined number of information bits
- H03M13/09—Error detection only, e.g. using cyclic redundancy check [CRC] codes or single parity bit
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M13/00—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
- H03M13/03—Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words
- H03M13/05—Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words using block codes, i.e. a predetermined number of check bits joined to a predetermined number of information bits
- H03M13/13—Linear codes
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M13/00—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
- H03M13/29—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes combining two or more codes or code structures, e.g. product codes, generalised product codes, concatenated codes, inner and outer codes
- H03M13/2906—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes combining two or more codes or code structures, e.g. product codes, generalised product codes, concatenated codes, inner and outer codes using block codes
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M13/00—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
- H03M13/63—Joint error correction and other techniques
- H03M13/6306—Error control coding in combination with Automatic Repeat reQuest [ARQ] and diversity transmission, e.g. coding schemes for the multiple transmission of the same information or the transmission of incremental redundancy
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0056—Systems characterized by the type of code used
- H04L1/0057—Block codes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0056—Systems characterized by the type of code used
- H04L1/0061—Error detection codes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
- H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
- H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
- H04L1/1812—Hybrid protocols; Hybrid automatic repeat request [HARQ]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
- H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
- H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
- H04L1/1829—Arrangements specially adapted for the receiver end
- H04L1/1861—Physical mapping arrangements
Definitions
- the present disclosure relates to a method and apparatus for transmitting data based on polar coding in a wireless communication system, and more particularly, to a method and apparatus for processing retransmission based on polar coding according to machine learning.
- a receiver measures a channel and performs decoding based on the channel measurement.
- a transmitter transmits relatively a large number of pilot signals, and overhead due to the pilot signals is not insignificant.
- performance optimization has not been achieved from the perspective of the whole system.
- NOMA non-orthogonal multiple access
- An object of the present disclosure is to provide a method and apparatus for processing retransmission based on polar coding according to machine learning.
- Another object of the present disclosure is to provide a method of effectively combining polar codes with hybrid automatic repeat and request (HARQ) based on machine learning to reduce the number of retransmissions and improve error correction performance.
- HARQ hybrid automatic repeat and request
- Another object of the present disclosure is to provide a method of simultaneously optimizing transmission of a pilot signal for channel measurement and transmission of polar codes based on machine learning to improve the performance of a communication system.
- Another object of the present disclosure is to provide a method of improving the performance of a non-orthogonal multiple access (NOMA) system based on machine learning.
- NOMA non-orthogonal multiple access
- a method of transmitting data based on polar coding in a wireless communication system may include: transmitting data including a plurality of information blocks, wherein each of the plurality of information blocks may include a corresponding cyclic redundancy check (CRC); receiving a hybrid automatic repeat request acknowledgement/negative acknowledgement (HARQ ACK/NACK) for the transmitted data; learning to retransmit the plurality of information blocks; and retransmitting the plurality of information blocks based on the HARQ ACK/NACK.
- CRC cyclic redundancy check
- HARQ ACK/NACK hybrid automatic repeat request acknowledgement/negative acknowledgement
- the learning may include: obtaining a current state s n ; obtaining actions to be applied to the current state s n ; and selecting an action for maximizing an expected reward value Q n+1 from among the actions, wherein the expected reward value Q n+1 may be obtained based on rewards R 1 , R 2 , . . . , R n respectively corresponding to states s 1 , s 2 , . . . , s n .
- the plurality of information blocks may be retransmitted based on the selected action.
- the expected reward value Q n+1 may be defined by the following equation based on a latest reward R n among the rewards R 1 , R 2 , . . . , R n , and a previous expected reward value Q n .
- Q n+1 ⁇ R n +(1 ⁇ ) Q n
- a learning rate ⁇ may be determined based on a range of channel variation.
- the action may include a first action of transmitting the plurality of information blocks without coding, a second action of coding and transmitting the plurality of information blocks, and a third action of coding and transmitting some of the plurality of information blocks and transmitting the remaining blocks without coding.
- Each of the rewards corresponding to the states may be obtained based on a cumulative number of bits of multiple information blocks transmitted up to now and the HARQ ACK/NACK, and the cumulative number of bits of multiple information blocks transmitted up to now and the HARQ ACK/NACK may be obtained based on the current state and the selected action.
- the expected reward value Q n+1 may be a weighted average of the rewards based on a learning rate.
- the learning rate may decrease monotonically as the learning progresses, or the learning rate may increase monotonically as the learning progresses.
- the expected reward value Q n+1 may be defined by the following equation based on the rewards R 1 , R 2 , . . . , R n .
- the expected reward value Q n+1 may be defined by the following equation based on a latest reward R n among the rewards R 1 , R 2 , . . . , R n and a previous expected reward value Q n .
- Q n+1 ⁇ n R n +(1 ⁇ ) Q n
- a learning rate ⁇ n may decrease monotonically as n increases.
- the learning rate ⁇ n may increase monotonically as n increases.
- an apparatus for transmitting data based on polar coding in a wireless communication system may include: a transceiver; a memory; and at least one processor connected to the transceiver and the memory.
- the memory may be configured to store instructions that, when executed, cause the at least one processor to perform operations including: transmitting data including a plurality of information blocks, wherein each of the plurality of information blocks may include a corresponding CRC; receiving a HARQ ACK/NACK for the transmitted data; learning to retransmit the plurality of information blocks; and retransmitting the plurality of information blocks based on the HARQ ACK/NACK.
- the learning may include: obtaining a current state s n ; obtaining actions to be applied to the current state s n ; and selecting an action for maximizing an expected reward value Q n+1 from among the actions, wherein the expected reward value Q n+1 may be obtained based on rewards R 1 , R 2 , . . . , R n , respectively corresponding to states s 1 , s 2 , . . . , s n .
- the plurality of information blocks may be retransmitted based on the selected action.
- the apparatus may be mounted on an autonomous driving apparatus configured to communicate with at least one of a mobile terminal, a base station, or an autonomous driving vehicle.
- retransmission based on polar coding may be processed by machine learning.
- the performance of a wireless communication system using polar coding may be improved by machine learning.
- the present disclosure may provide a polar code-based HARQ method for performance improvement when polar codes are used for hybrid automatic repeat request (HARQ) based on machine learning.
- HARQ hybrid automatic repeat request
- the present disclosure may provide a method of improving performance by combining polar coding with a non-orthogonal multiple access (NOMA) system, which has recently been spotlighted as a standard technology for 5G wireless communication systems, based on machine learning.
- NOMA non-orthogonal multiple access
- maximum channel estimation and decoding effects may be obtained by adding an optimal amount of pilot signals to a polar code.
- FIG. 1 A illustrates a communication system applied to the present disclosure.
- FIG. 1 B illustrates wireless devices applicable to the present disclosure.
- FIG. 1 C illustrates other examples of wireless devices applicable to the present disclosure.
- FIG. 2 is a diagram illustrating a frame structure of a new radio access technology (new RAT or NR).
- FIG. 3 illustrates a resource grid of NR.
- FIG. 4 is a diagram for explaining a channel coding method according to the present disclosure.
- FIGS. 5 and 6 are diagrams for explaining a modulation method according to the present disclosure.
- FIG. 7 is a diagram for explaining reinforcement learning.
- FIG. 8 is a diagram for explaining a backpropagation method in a neural network.
- FIG. 9 is a diagram for explaining a prediction method in an artificial neural network.
- FIG. 10 A is a diagram for explaining a method of operating a recurrent neural network.
- FIG. 10 B is a diagram for explaining an operating method of long short-term memory (LSTM).
- FIG. 11 is a diagram for explaining a method of adding a cyclic redundancy check (CRC) to a polar code and a method of encoding and decoding a polar code using multiple CRCs.
- CRC cyclic redundancy check
- FIG. 12 is a diagram for explaining a retransmission method for polar codes.
- FIG. 13 is a diagram for explaining a NOMA system model with two users.
- FIG. 14 shows NOMA system method 1 (method 1) based on polar coding.
- FIG. 15 shows NOMA system method 2 (method 2) based on polar coding.
- FIG. 16 is a diagram conceptually representing FIGS. 14 and 15 .
- FIG. 17 shows a retransmission method in a NOMA system using only one CRC for each layer.
- FIG. 18 is a diagram for explaining a method of actively changing the value of c depending on channel environments according to the present disclosure.
- FIG. 19 is a diagram for explaining a method of actively changing the value of c depending on channel environments (e.g., channel coherence time) according to the present disclosure.
- FIGS. 20 and 21 are diagrams for explaining a method of actively changing the value of depending on channel environments according to the present disclosure.
- FIG. 22 is a diagram for explaining a method of actively changing the value of a depending on channel environments according to the present disclosure.
- FIG. 23 is a diagram for explaining optimization of a HARQ procedure and a system model.
- FIG. 24 shows a case in which one codeword includes four information blocks.
- FIG. 25 is a diagram for explaining a retransmission method in a NOMA system using one CRC in one layer.
- FIG. 26 is a diagram for explaining a retransmission method in a NOMA system using a plurality of CRCs in one layer.
- FIGS. 27 A and 27 B are diagrams for explaining a method of combining a systematic polar code with channel measurement according to the present disclosure.
- FIG. 28 A is a diagram for explaining a method of combining a non-systematic polar code with channel measurement according to the present disclosure
- FIG. 28 B is a diagram for explaining a method of combining a non-systematic polar code with channel measurement where permutation is used to arrange pilot signals at a regular interval according to the present disclosure.
- FIG. 29 is a diagram for explaining a method of transmitting data based on polar coding according to the present disclosure.
- FIG. 30 is a diagram for explaining a method of receiving data based on polar coding according to the present disclosure.
- a terminal is a common name of such a mobile or fixed user stage device as a user equipment (UE), a mobile station (MS), an advanced mobile station (AMS) and the like.
- a base station (BS) is a common name of such a random node of a network stage communicating with a terminal as a Node B (NB), an eNode B (eNB), an access point (AP) and the like.
- a user equipment In a mobile communication system, a user equipment is able to receive information in downlink and is able to transmit information in uplink as well.
- Information transmitted or received by the user equipment node may include various kinds of data and control information.
- various physical channels may exist.
- CDMA Code Division Multiple Access
- FDMA Frequency Division Multiple Access
- TDMA Time Division Multiple Access
- OFDMA Orthogonal Frequency Division Multiple Access
- SC-FDMA Single Carrier Frequency Division Multiple Access
- CDMA may be implemented as a radio technology such as Universal Terrestrial Radio Access (UTRA) or CDMA2000.
- TDMA may be implemented as a radio technology such as Global System for Mobile communications (GSM)/General packet Radio Service (GPRS)/Enhanced Data Rates for GSM Evolution (EDGE).
- GSM Global System for Mobile communications
- GPRS General packet Radio Service
- EDGE Enhanced Data Rates for GSM Evolution
- OFDMA may be implemented as a radio technology such as IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, Evolved UTRA (E-UTRA), etc.
- UTRA is a part of Universal Mobile Telecommunications System (UMTS).
- 3GPP LTE is a part of Evolved UMTS (E-UMTS) using E-UTRA, adopting OFDMA for DL and SC-FDMA for UL.
- LTE-Advanced (LTE-A) is an evolution of 3GPP LTE.
- FIG. 1 A illustrates a communication system 1 applied to the present disclosure.
- the communication system 1 applied to the present disclosure includes wireless devices, BSs, and a network.
- the wireless devices refer to devices performing communication by radio access technology (RAT) (e.g., 5G New RAT (NR) or LTE), which may also be called communication/radio/5G devices.
- RAT radio access technology
- the wireless devices may include, but no limited to, a robot 100 a , vehicles 100 b - 1 and 100 b - 2 , an extended reality (XR) device 100 c , a hand-held device 100 d , a home appliance 100 e , an IoT device 100 f , and an artificial intelligence (AI) device/server 400 .
- RAT radio access technology
- NR 5G New RAT
- LTE Long Term Evolution
- the wireless devices may include, but no limited to, a robot 100 a , vehicles 100 b - 1 and 100 b - 2 , an extended reality (XR) device 100 c , a hand-held device 100 d
- the vehicles may include a vehicle equipped with a wireless communication function, an autonomous driving vehicle, and a vehicle capable of performing vehicle-to-vehicle (V2V) communication.
- the vehicles may include an unmanned aerial vehicle (UAV) (e.g., a drone).
- UAV unmanned aerial vehicle
- the XR device may include an augmented reality (AR)/virtual reality (VR)/mixed reality (MR) device, and may be implemented in the form of a head-mounted device (HMD), a head-up display (HUD) mounted in a vehicle, a television (TV), a smartphone, a computer, a wearable device, a home appliance, a digital signage, a vehicle, a robot, and so on.
- AR augmented reality
- VR virtual reality
- MR mixeded reality
- the hand-held device may include a smartphone, a smartpad, a wearable device (e.g., a smartwatch or smart glasses), and a computer (e.g., a laptop).
- the home appliance may include a TV, a refrigerator, and a washing machine.
- the IoT device may include a sensor and a smart meter.
- the BSs and the network may be implemented as wireless devices, and a specific wireless device 200 a may operate as a BS/network node for other wireless devices.
- the wireless devices 100 a to 100 f may be connected to the network 300 via the BSs 200 .
- An AI technology may be applied to the wireless devices 100 a to 100 f , and the wireless devices 100 a to 100 f may be connected to the AI server 400 via the network 300 .
- the network 300 may be configured by using a 3G network, a 4G (e.g., LTE) network, or a 5G (e.g., NR) network.
- the wireless devices 100 a to 100 f may communicate with each other through the BSs 200 /network 300
- the wireless devices 100 a to 100 f may perform direct communication (e.g., sidelink communication) with each other without intervention of the BSs/network.
- the vehicles 100 b - 1 and 100 b - 2 may perform direct communication (e.g. V2V/vehicle-to-everything (V2X) communication).
- the IoT device e.g., a sensor
- the IoT device may perform direct communication with other IoT devices (e.g., sensors) or other wireless devices 100 a to 100 f.
- Wireless communication/connections 150 a , 150 b , or 150 c may be established between the wireless devices 100 a to 100 f and the BSs 200 , or between the BSs 200 .
- the wireless communication/connections may be established through various RATs (e.g., 5G NR) such as UL/DL communication 150 a , sidelink communication 150 b (or, D2D communication), or inter-BS communication 150 c (e.g. relay, integrated access backhaul (IAB)).
- a wireless device and a BS/a wireless devices, and BSs may transmit/receive radio signals to/from each other through the wireless communication/connections 150 a , 150 b , and 150 c .
- various configuration information configuring processes various signal processing processes (e.g., channel encoding/decoding, modulation/demodulation, and resource mapping/demapping), and resource allocating processes, for transmitting/receiving radio signals, may be performed based on the various proposals of the present disclosure.
- various signal processing processes e.g., channel encoding/decoding, modulation/demodulation, and resource mapping/demapping
- resource allocating processes for transmitting/receiving radio signals
- FIG. 1 B illustrates wireless devices applicable to the present disclosure.
- a first wireless device 100 and a second wireless device 200 may transmit radio signals through a variety of RATs (e.g., LTE and NR).
- ⁇ the first wireless device 100 and the second wireless device 200 ⁇ may correspond to ⁇ the wireless devices 100 a to 100 f and the BSs 200 ⁇ and/or ⁇ the wireless devices 100 a to 100 f and the wireless devices 100 a to 100 f ⁇ of FIG. 1 A .
- the first wireless device 100 may include at least one processor 102 and at least one memory 104 , and may further include at least one transceiver 106 and/or at least one antenna 108 .
- the processor 102 may control the memory 104 and/or the transceiver 106 and may be configured to implement the descriptions, functions, procedures, proposals, methods, and/or operational flowcharts disclosed in this document.
- the processor 102 may process information within the memory 104 to generate first information/signal and then transmit a radio signal including the first information/signal through the transceiver 106 .
- the processor 102 may receive a radio signal including second information/signal through the transceiver 106 and then store information obtained by processing the second information/signal in the memory 104 .
- the memory 104 may be coupled to the processor 102 and store various types of information related to operations of the processor 102 .
- the memory 104 may store software code including commands for performing a part or all of processes controlled by the processor 102 or for performing the descriptions, functions, procedures, proposals, methods, and/or operational flowcharts disclosed in this document.
- the processor 102 and the memory 104 may be a part of a communication modem/circuit/chip designed to implement an RAT (e.g., LTE or NR).
- the transceiver 106 may be coupled to the processor 102 and transmit and/or receive radio signals through the at least one antenna 108 .
- the transceiver 106 may include a transmitter and/or a receiver.
- the transceiver 106 may be interchangeably used with an RF unit.
- a wireless device may refer to a communication modem/circuit/chip.
- the second wireless device 200 may include at least one processor 202 and at least one memory 204 , and may further include at least one transceiver 206 and/or at least one antenna 208 .
- the processor 202 may control the memory 204 and/or the transceiver 206 and may be configured to implement the descriptions, functions, procedures, proposals, methods, and/or operational flowcharts disclosed in this document.
- the processor 202 may process information within the memory 204 to generate third information/signal and then transmit a radio signal including the third information/signal through the transceiver 206 .
- the processor 202 may receive a radio signal including fourth information/signal through the transceiver 206 and then store information obtained by processing the fourth information/signal in the memory 204 .
- the memory 204 may be coupled to the processor 202 and store various types of information related to operations of the processor 202 .
- the memory 204 may store software code including commands for performing a part or all of processes controlled by the processor 202 or for performing the descriptions, functions, procedures, proposals, methods, and/or operational flowcharts disclosed in this document.
- the processor 202 and the memory 204 may be a part of a communication modem/circuit/chip designed to implement an RAT (e.g., LTE or NR).
- the transceiver 206 may be coupled to the processor 202 and transmit and/or receive radio signals through the at least one antenna 208 .
- the transceiver 206 may include a transmitter and/or a receiver.
- the transceiver 206 may be interchangeably used with an RF unit.
- a wireless device may refer to a communication modem/circuit/chip.
- One or more protocol layers may be implemented by, but not limited to, one or more processors 102 and 202 .
- the one or more processors 102 and 202 may implement one or more layers (e.g., functional layers such as PHY, MAC, RLC, PDCP, RRC, and SDAP).
- the one or more processors 102 and 202 may generate one or more protocol data units (PDUs) and/or one or more service data units (SDUs) according to the descriptions, functions, procedures, proposals, methods, and/or operational flowcharts disclosed in this document.
- PDUs protocol data units
- SDUs service data units
- the one or more processors 102 and 202 may generate messages, control information, data, or information according to the descriptions, functions, procedures, proposals, methods, and/or operational flowcharts disclosed in this document.
- the one or more processors 102 and 202 may generate signals (e.g., baseband signals) including PDUs, SDUs, messages, control information, data, or information according to the descriptions, functions, procedures, proposals, methods, and/or operational flowcharts disclosed in this document and provide the generated signals to the one or more transceivers 106 and 206 .
- the one or more processors 102 and 202 may receive the signals (e.g., baseband signals) from the one or more transceivers 106 and 206 and acquire the PDUs, SDUs, messages, control information, data, or information according to the descriptions, functions, procedures, proposals, methods, and/or operational flowcharts disclosed in this document.
- signals e.g., baseband signals
- the one or more processors 102 and 202 may be referred to as controllers, microcontrollers, microprocessors, or microcomputers.
- the one or more processors 102 and 202 may be implemented in hardware, firmware, software, or a combination thereof.
- ASICs application specific integrated circuits
- DSPs digital signal processors
- DSPDs digital signal processing devices
- PLDs programmable logic devices
- FPGAs field programmable gate arrays
- Firmware or software configured to perform the descriptions, functions, procedures, proposals, methods, and/or operational flowcharts disclosed in this document may be included in the one or more processors 102 and 202 , or may be stored in the one or more memories 104 and 204 and executed by the one or more processors 102 and 202 .
- the descriptions, functions, procedures, proposals, methods, and/or operational flowcharts disclosed in this document may be implemented as code, instructions, and/or a set of instructions in firmware or software.
- the one or more memories 104 and 204 may be coupled to the one or more processors 102 and 202 and store various types of data, signals, messages, information, programs, code, instructions, and/or commands.
- the one or more memories 104 and 204 may be configured as read-only memories (ROMs), random access memories (RAMs), electrically erasable programmable read-only memories (EPROMs), flash memories, hard drives, registers, cash memories, computer-readable storage media, and/or combinations thereof.
- the one or more memories 104 and 204 may be located at the interior and/or exterior of the one or more processors 102 and 202 .
- the one or more memories 104 and 204 may be coupled to the one or more processors 102 and 202 through various technologies such as wired or wireless connection.
- the one or more transceivers 106 and 206 may transmit user data, control information, and/or radio signals/channels, mentioned in the methods and/or operational flowcharts of this document, to one or more other devices.
- the one or more transceivers 106 and 206 may receive user data, control information, and/or radio signals/channels, mentioned in the descriptions, functions, procedures, proposals, methods, and/or operational flowcharts disclosed in this document, from one or more other devices.
- the one or more transceivers 106 and 206 may be coupled to the one or more processors 102 and 202 and transmit and receive radio signals.
- the one or more processors 102 and 202 may control the one or more transceivers 106 and 206 to transmit user data, control information, or radio signals to one or more other devices.
- the one or more processors 102 and 202 may control the one or more transceivers 106 and 206 to receive user data, control information, or radio signals from one or more other devices.
- the one or more transceivers 106 and 206 may be coupled to the one or more antennas 108 and 208 and configured to transmit and receive user data, control information, and/or radio signals/channels, mentioned in the descriptions, functions, procedures, proposals, methods, and/or operational flowcharts disclosed in this document, through the one or more antennas 108 and 208 .
- the one or more antennas may be a plurality of physical antennas or a plurality of logical antennas (e.g., antenna ports).
- the one or more transceivers 106 and 206 may convert received radio signals/channels etc. from RF band signals into baseband signals in order to process received user data, control information, radio signals/channels, etc. using the one or more processors 102 and 202 .
- the one or more transceivers 106 and 206 may convert the user data, control information, radio signals/channels, etc. processed using the one or more processors 102 and 202 from the base band signals into the RF band signals.
- the one or more transceivers 106 and 206 may include (analog) oscillators and/or filters.
- FIG. 1 C illustrates another example of wireless devices applied to the present disclosure.
- the wireless devices may be implemented in various forms according to use-cases/services (refer to FIG. 1 A ).
- wireless devices 100 and 200 may correspond to the wireless devices 100 and 200 of FIG. 1 B and may be configured as various elements, components, units/portions, and/or modules.
- each of the wireless devices 100 and 200 may include a communication unit 110 , a control unit 120 , a memory unit 130 , and additional components 140 .
- the communication unit may include a communication circuit 112 and transceiver(s) 114 .
- the communication circuit 112 may include the one or more processors 102 and 202 and/or the one or more memories 104 and 204 of FIG. 2 B .
- the transceiver(s) 114 may include the one or more transceivers 106 and 206 and/or the one or more antennas 108 and 208 of FIG. 2 B .
- the control unit 120 is electrically coupled to the communication unit 110 , the memory unit 130 , and the additional components 140 and provides overall control to operations of the wireless devices.
- the control unit 120 may control an electric/mechanical operation of the wireless device based on programs/code/commands/information stored in the memory unit 130 .
- the control unit 120 may transmit the information stored in the memory unit 130 to the outside (e.g., other communication devices) via the communication unit 110 through a wireless/wired interface or store, in the memory unit 130 , information received through the wireless/wired interface from the outside (e.g., other communication devices) via the communication unit 110 .
- the additional components 140 may be configured in various manners according to the types of wireless devices.
- the additional components 140 may include at least one of a power unit/battery, an input/output (I/O) unit, a driver, and a computing unit.
- the wireless device may be configured as, but not limited to, the robot ( 100 a of FIG. 1 A ), the vehicles ( 100 b - 1 and 100 b - 2 of FIG. 1 A ), the XR device ( 100 c of FIG. 1 A ), the hand-held device ( 100 d of FIG. 1 A ), the home appliance ( 100 e of FIG. 1 A ), the IoT device ( 100 f of FIG.
- the wireless device may be mobile or fixed according to a use-case/service.
- all of the various elements, components, units/portions, and/or modules in the wireless devices 100 and 200 may be coupled to each other through a wired interface or at least a part thereof may be wirelessly coupled to each other through the communication unit 110 .
- the control unit 120 and the communication unit 110 may be coupled by wire, and the control unit 120 and first units (e.g., 130 and 140 ) may be wirelessly coupled through the communication unit 110 .
- Each element, component, unit/portion, and/or module within the wireless devices 100 and 200 may further include one or more elements.
- the control unit 120 may be configured as a set of one or more processors.
- control unit 120 may be configured as a set of a communication control processor, an application processor, an electronic control unit (ECU), a graphical processing unit, and a memory control processor.
- memory unit 130 may be configured as a random access memory (RAM), a dynamic RAM (DRAM), a read only memory (ROM), a flash memory, a volatile memory, a non-volatile memory, and/or a combination thereof.
- a device for performing channel coding based on polar coding may include a transceiver, a memory, and at least one processor connected to the transceiver and memory.
- the memory may be configured to store instructions that, when executed by the at least one processor, cause the at least one processor to perform operations.
- FIG. 2 is a diagram illustrating a frame structure in NR.
- the NR system may support multiple numerologies.
- a numerology may be defined by a subcarrier spacing (SCS) and a cyclic prefix (CP) overhead.
- Multiple SCSs may be derived by scaling a default SCS by an integer N (or ⁇ ). Further, even though it is assumed that a very small SCS is not used in a very high carrier frequency, a numerology to be used may be selected independently of a frequency band. Further, the NR system may support various frame structures according to multiple numerologies.
- OFDM numerologies and frame structures which may be considered for the NR system.
- Multiple OFDM numerologies supported by the NR system may be defined as listed in Table 1.
- the NR system supports multiple numerologies (e.g., SCSs) to support various 5G services. For example, in an SCS of 15 kHz, the NR system supports a wide area in conventional cellular bands. In an SCS of 30/60 kHz, the NR system supports a dense urban environment, low latency, and wide carrier bandwidth. In an SCS of 60 kHz or above, the NR system supports a bandwidth higher than 24.25 GHz to overcome phase noise.
- SCSs numerologies
- NR frequency bands are divided into two frequency ranges: frequency range 1 (FR1) and frequency range 2 (FR2).
- FR1 covers sub-6 GHz frequency bands
- FR2 covers frequency bands above 6 GHz, i.e., bands in the millimeter wavelength (mmWave).
- Table 2 shows the definitions of the NR frequency ranges.
- slots are numbered with n ⁇ s ⁇ 0, . . . , N slot, ⁇ subframe ⁇ 1 ⁇ in an increasing order in a subframe, and with n ⁇ s,f ⁇ 0, . . . , N slot, ⁇ frame ⁇ 1 ⁇ in an increasing order in a radio frame.
- One slot includes N ⁇ symb consecutive OFDM symbols, and N ⁇ symb depends on a used numerology and slot configuration.
- the start of a slot n s ⁇ in a subframe is aligned in time with the start of an OFDM symbol n ⁇ s *N ⁇ symb in the same subframe.
- All UEs are not capable of simultaneous transmission and reception, which implies that all OFDM symbols of a DL slot or a UL slot may not be used.
- Table 3 lists the number N slot symb of symbols per slot, the number N frame, ⁇ slot of slots per frame, and the number of slots per subframe, for each SCS in a normal CP case
- Table 4 lists the number of symbols per slot, the number of slots per frame, and the number of slots per subframe, for each SCS in an extended CP case.
- One subframe ⁇ 1, 2, 4 ⁇ slots in FIG. 2 which is exemplary, and the number of slot(s) which may be included in one subframe is defined as listed in Table 2.
- a mini-slot may include 2, 4 or 7 symbols, fewer symbols than 2, or more symbols than 7.
- an antenna port, a resource grid, a resource element, a resource block, a carrier part, etc. may be considered as physical resources.
- the physical resources considerable in the NR system will be described in detail.
- an antenna port may be defined such that a channel conveying symbols on the antenna port is capable of being inferred from a channel conveying other symbols on the same antenna port.
- the two antenna ports may be said to be in quasi co-located or quasi co-location (QC/QCL) relationship.
- the large-scale properties may include at least one of the following parameters: delay spread, Doppler spread, frequency shift, average received power, received timing, average delay, and spatial reception (Rx).
- the spatial Rx parameter refer to a spatial (Rx) channel characteristic parameter such as angle of arrival.
- FIG. 3 illustrates a resource grid in the NR system.
- a resource grid includes N RB ⁇ N sc RB subcarriers in the time domain, and one subframe 14 ⁇ 2 ⁇ OFDM symbols, which is exemplary and thus should not be construed as limiting the disclosure.
- a transmitted signal is described by one or more resource grids including N RB ⁇ N sc RB subcarriers and 2 ⁇ N symb ( ⁇ ) OFDM symbols, where N RB ⁇ ⁇ N RB max, ⁇ ⁇ N RB max, ⁇ represents a maximum transmission bandwidth, which may be different for UL and DL as well as according to numerologies.
- one resource grid may be configured for each neurology ⁇ and each antenna port p, as illustrated in FIG.
- An RE (k, l ) for the numerology ⁇ and the antenna port p corresponds to a complex value a k,l (p, ⁇ ) .
- the indexes p and ⁇ may be dropped, and as a result, the complex value may be a k,l (p) or a k, l .
- the UE may be configured to operate in a part of the frequency BW of a cell (hereinafter referred to as a bandwidth part (BWP)).
- BWP bandwidth part
- resource blocks may be divided into physical resource blocks defined within the BWP and common resource blocks numbered from 0 upward in the frequency domain for an SCS configuration ⁇ .
- Point A is obtained as follows.
- the center of subcarrier 0 of common resource block 0 for the SCS configuration ⁇ coincides with point A, which act as the reference point for resource grids.
- the relationship between a common resource block number n ⁇ CRB in the frequency domain and a resource elements (k,l) for the SCS configuration ⁇ is given by Equation 1.
- Physical resource blocks are numbered from 0 to N size BWP,i-1 within the BWP, where i is the number of the BWP.
- the relationship between a physical resource block n PRB and a common resource block n CRB in BWP i is given by Equation 2.
- n CRB n PRB +N BWP,i start [Equation 2]
- N start BWP,i is a common resource block where the BWP starts relative to common resource block 0.
- FIG. 4 is a diagram for explaining a channel coding method according to the present disclosure.
- the transport block is divided into code blocks, each of which has a size less than or equal to a predetermined value.
- the code block may have a maximum size of 6144 bits.
- the code block may have a maximum size of 8448 bits (in base graph 1) or 3840 bits (in base graph 2).
- the code block may have a minimum size of 32 bits and a maximum size of 8192 bits.
- the code block may be subdivided into sub-blocks.
- an input bit sequence ( 265 ) (c r0 , c r1 , . . . , cr (Kr ⁇ 1) ) is interleaved, the interleaved input bit sequence (not shown in the drawing) (c′ r0 , c′ r1 , . . . , c′ r(Kr ⁇ 1) ) may be encoded based on polar codes.
- the encoded bit sequence ( 270 ) (d r0 , d r1 , . . . , d r(Nr ⁇ 1) ) may be rate matched.
- the rate matching of the encoded bit sequence ( 270 ) may include: subdividing the encoded bit sequence into sub-blocks; interleaving each of the sub-blocks; performing bit selection for each of the interleaved sub-blocks; and interleaving coded bits again.
- the bit selection for each of the interleaved sub-blocks may include repeating, puncturing, or shortening some bits.
- the channel coding method may include attaching a cyclic redundancy check (CRC) code to a transport block (S 205 ); dividing the transport block into code blocks (S 210 ); encoding the divided code blocks (S 215 ); perform rate matching of the encoded code blocks (S 220 ); and concatenating the rate-matched code blocks (S 225 ).
- CRC cyclic redundancy check
- step S 205 party bits with a length of L are attached to the transport block ( 255 ) (a 0 , . . . , a A-1 ).
- the length L may be any one of 6, 11, 16, and 24.
- cyclic generator polynomials are used to generate party bits.
- scrambling operation may be applied to output bits ( 260 ) (b 0 , . . . , b B ⁇ 1 ), which depend on the CRC attachment, with a radio network temporary identifier (RNTI).
- Exclusive OR (EOR) operation may be applied between a scrambling sequence and corresponding bits based on the scrambling operation.
- the output bits ( 260 ) (b 0 , . . . , b B ⁇ 1 ) depending on the CRC attachment may be segmented into code blocks ( 265 ) according to code block sizes. This is called code block segmentation.
- the code block sizes are determined by channel coding methods. A code block size suitable for each channel coding method may be determined theoretically or experimentally.
- the segmented code blocks ( 265 ) (c r0 , . . . , c r(Kr ⁇ 1) ) may be encoded as encoded bits ( 270 ) (d r0 , . . . , d r(Nr ⁇ 1) ), respectively.
- the encoded bits ( 270 ) (d r0 , . . . , d r(Nr ⁇ 1) ) are generated by applying channel coding to the code blocks ( 265 ) (c r0 , . . . , c r(Kr ⁇ 1) ) (S 215 ).
- the generated encoded bits ( 270 ) may be rate-matched by shortening and puncturing.
- the encoded bits ( 270 ) may be rate-matched by sub-block interleaving, bit selection, and/or interleaving. That is, the encoded bits ( 270 ) (d r0 , . . .
- interleaving may refer to a process for changing a sequence of bits and reduce the occurrence of errors.
- the interleaving is designed in consideration of efficient de-interleaving.
- Sub-block interleaving may mean a process for dividing a code block into a plurality of sub-blocks (e.g., 32 sub-blocks) and allocating bits based on the sub-block interleaving.
- the bit selection may mean a process for increasing a bit sequence by repeating bits based on the number of bits to be rate-matched or decreasing the bit sequence based on shortening, puncturing, etc.
- the interleaving may mean a process for interleaving encoded bits after the bit selection.
- the rate matching may include the bit selection and interleaving.
- the sub-block interleaving is not mandatory.
- code block concatenation is applied to concatenate the code blocks ( 275 ) so that a codeword ( 285 ) (g 0 , . . . , g G ⁇ 1 ) may be generated (S 225 ).
- the generated codeword 280 may be equivalent to one transport block.
- FIGS. 5 and 6 are diagrams for explaining a modulation method according to the present disclosure.
- one or more codewords are input and scrambled (S 305 and S 405 ).
- scrambling may be performed based on EOR operation between an input bit sequence and a predetermined bit sequence.
- the scrambled bits are modulated (S 310 and S 410 ), and the modulated symbols are mapped to layers (S 315 and S 415 ).
- the symbols mapped to the layers are precoded for antenna port mapping.
- the precoded symbols are mapped to resource elements (S 325 and S 425 ).
- the mapped symbols are generated as OFDM signals (S 330 and S 430 ) and transmitted through antennas.
- the receiver measures a channel and performs decoding based on the channel measurement.
- the transmitter transmits relatively a large number of pilot signals, and overhead due to the pilot signals is not negligible.
- performance optimization has not been achieved from the perspective of the whole system.
- the present disclosure proposes a polar code based HARQ method for minimizing the number of times of retransmission and improving the performance of error correction by efficiently combining polar codes with HARQ.
- the present disclosure proposes a method of improving the performance of a communication system by simultaneously optimizing pilot signal transmission for channel measurement and polar code transmission at a receiver.
- FIGS. 7 A and 7 B are diagrams for showing performance improvement in polar coding based channel coding according to the present disclosure when pilot signals are applied to codewords of a polar code, compared to when puncturing is applied.
- a polar code is a linear block error correction code.
- the code structure is based on multiple recursive concatenation of short kernel codes which transform a physical channel into virtual outer channels.
- the polar coding method according to the present disclosure is advantageous in that fast decoding is enabled because the generator matrix is easily determined and a reverse matrix thereof is rapidly computed due to the feature of the polar code generator matrix.
- the present disclosure may solve various communication problems based on a multi-armed bandit algorithm, Q-learning, and deep Q-network (DQN).
- DQN deep Q-network
- Such a method has the following main features: not only prior knowledge about channel or system environments but also any mathematical modeling therefor are not required, but optimal actions are taken and rewards are given for the actions, thereby learning surrounding environments and eventually choosing the best action.
- the present disclosure proposes effective methods for solving retransmission problems for polar codes, retransmission problems in non-orthogonal multiple access (NOMA) systems, HARQ problems, and pilot insertion problems for polar codes.
- NOMA non-orthogonal multiple access
- Reinforcement learning is a type of machine learning, and among supervised learning and unsupervised learning, the reinforcement learning may be classified as the unsupervised learning.
- the main feature of reinforcement learning is that it does not require any prior knowledge or mathematical modeling for environments. In communications, many assumptions are made to attempt to solve problems in communication systems based on mathematical modeling. In this case, if even one of the assumptions is invalid, no algorithm may not work in practice. On the contrary, in reinforcement learning, no assumptions are made in advance. Instead, environments are learned based on rewards given for actions performed by an agent under the environments, and the best action is chosen according to learning results. This feature of reinforcement learning is also very useful to optimize communication systems in real environments.
- the present disclosure proposes methods for solving communication problems based on multi-armed bandit (MAB), Q-learning, and DQN algorithms. In addition, the present disclosure proposes methods for changing various parameters of a reinforcement learning algorithm effectively and dynamically according to change in channel environments.
- the purpose of a digital communication system is to transfer data (bits) from a transmitter to a receiver without errors.
- Multiple techniques are currently used to achieve this purpose, and polar codes are one of the techniques.
- the polar codes are classified into non-systematic polar codes (reference 1) and systematic polar codes (reference 2).
- the present disclosure proposes retransmission methods in the NOMA communication system.
- one or more users may transmit data in the same frequency band at the same time.
- one or more users may transmit data in frequency bands at the same time.
- the receiver decodes data by using a successive interference cancellation (SIC) decoder.
- SIC successive interference cancellation
- the NOMA system may increase the overall transmission rate from the system point of view.
- Even in the NOMA system when initial transmission fails, retransmission needs to be performed. Thus, it is very important to determine which retransmission method provides the best performance among possible various methods. However, as described above, it is very difficult to solve such a problem mathematically or analytically.
- the present disclosure proposes methods for efficiently transmitting a pilot signal for channel measurement.
- the receiver requires channel information for decoding systematic or non-systematic polar codes.
- One method is to separately transmit a pilot signal to measure channel information, but a more efficient method is to transmit a pilot signal as a partial polar codeword.
- a more efficient method is to transmit a pilot signal as a partial polar codeword.
- the present disclosure may provide the following methods based on the MAB, Q-learning, and DQN algorithms.
- an optimal retransmission method may be determined based on the MAB algorithm.
- an optimal retransmission method may be determined based on the MAB algorithm.
- various parameter values may be actively adjusted depending on change in the wireless channel.
- Q-learning may be used to solve general HARQ problems.
- Q-learning may be used to apply HARQ to the polar code communication system.
- Q-learning may be used to apply HARQ to the NOMA communication system.
- various parameter values may be actively adjusted depending on change in the radio channel.
- DQN may be used to solve general HARQ problems.
- DQN may be used to transmit an optimal pilot signal in the polar code.
- DQN may be used to apply HARQ to the polar code communication system.
- DQN may be used to apply HARQ to the NOMA communication system.
- various parameter values may be actively adjusted depending on change in the radio channel.
- a MAB problem (or K-armed bandit problem) is related to how a fixed and limited set of resources are allocated between options to be computed. The characteristics of each option are known only at the time to be allocated.
- the MAB problem may correspond to a reinforcement learning problem that exemplifies the exploration-exploitation tradeoff dilemma.
- the MAB algorithm may relate to statistical scheduling.
- Reinforcement learning is a type of machine learning, specifically, may be classified as unsupervised learning.
- FIG. 7 is a diagram for explaining reinforcement learning.
- the agent obtains information (e.g., state) about the environment and determines an action. New information and rewards may be obtained from the environment changed by the determined action.
- information e.g., state
- New information and rewards may be obtained from the environment changed by the determined action.
- FIG. 8 is a diagram for explaining a backpropagation method in a neural network.
- backpropagation may be performed.
- o k of the input layer may simply be an input to the network, x k .
- o j may be defined as shown in Equation 5.
- the error function E may satisfy Equation 6.
- FIG. 9 is a diagram for explaining a prediction method in an artificial neural network.
- An artificial neural network includes an input layer composed of first input data, an output layer composed of last output data, and a hidden layer as an intermediate layer for calculating output data from input data. There are one or more hidden layers, and an artificial neural network including two or more hidden layers is called a deep neural network (DNN). The actual operation is performed at nodes in each floor, and each node may calculate based on the output values of other nodes connected by connection lines.
- DNN deep neural network
- input data do not affect each other, and nodes belonging to the same layer do not affect each other.
- Each layer exchanges data as input or output values only with nodes at adjacent layers, i.e., upper or lower layers.
- FIG. 9 shows that all nodes in different layers are connected to each other by connection lines, but connection lines between nodes belonging to adjacent layers may be omitted if necessary. However, when there is no connection line, processing may be performed by setting a weight of 0 for a corresponding input value.
- an input value may be predicted from the results during learning.
- input and output values do not have a one-to-one correspondence, it is difficult to restore the original input layer from the output layer as it is.
- the prediction of the artificial neural network may be considered to be inaccurate. Therefore, learning may be trained by changing prediction coefficients so that the input data calculated under constraints become similar to the first input data.
- FIG. 10 A is a diagram for explaining a method of operating a recurrent neural network.
- an output value a 0 may be predicted only from x 0 , unlike the artificial neural network of FIG. 9 .
- an output value b 0 may be calculated based thereon, and b 0 may be reused to predict a 1 .
- the artificial neural network of FIG. 9 it is assumed that multiple input data are simultaneously input. If the input data are time series data, prediction may be allowed only after all data are input. Thus, output values may be calculated by an RNN method to process the time series data. In addition, the RNN may be applied to the Markov decision process (MDP).
- MDP Markov decision process
- the MDP provides a reasonable format for plans and actions in the face of uncertainty.
- the MDP may have various definitions.
- the definitions of the MDP may be treated equally to variants in question.
- the MDP may include states, an initial state distribution, actions, state transition distributions, a discount factor, and a reward function.
- an event may be processed as follows. First, the event may start at an initial state s0 from the initial state distribution. An action a t may be selected at a time t, and then a state s t may transition to a state s t+1 based on the state transition distributions. That is, by repeatedly selecting actions a 0 , a 1 , a 2 , states s 1 , s 2 , s 3 , . . . may be obtained. Assuming that a discount factor ⁇ is the same for each step, a reward may be defined by R(s 0 )+ ⁇ *R(s 1 )+ ⁇ 2 *R(s 2 )+ ⁇ 3 *R(s 3 )+ . . . . Although the above example shows that the reward depends on only states, but the reward may depend on not only the states but actions. That is, the reward may be defined by R(s t , a t ).
- learning may be trained in the same way as shown in FIG. 9 .
- actual prediction is performed based on the trained learning, it may be efficient to use the RNN shown in FIG. 10 A .
- FIG. 10 B is a diagram for explaining an operating method of long short-term memory (LSTM).
- LSTM is a type of RNN method that predicts results based on forgetting gates instead of using weights of the RNN.
- old data may be reduced according to the weights, so that the value thereof may become zero after a certain stage. Thus, the old data may be no longer reflected, regardless of the weights.
- the present disclosure proposes methods for solving retransmission problems for polar codes and retransmission problems in NOMA systems.
- the present disclosure proposes methods of efficiently changing the parameters of the MAB algorithm according to the channel characteristics.
- FIG. 11 is a diagram for explaining a method for adding a CRC to a polar code and a method for encoding and decoding a polar code using multiple CRCs.
- the codeword of a polar code may be divided into a plurality of information blocks, and a CRC may be added to each of the plurality of information blocks and then transmitted. Since the CRC is added to each of the plurality of information blocks, retransmission may be performed for each of the plurality of information blocks.
- a CRC is usually added to each code block corresponding to a codeword (S 205 ).
- a CRC added to each information block, which is obtained by dividing a code block, is different from the CRC of step S 205 of FIG. 4 in that it is included in an internal processor for polar coding.
- FIG. 12 is a diagram for explaining a retransmission method for polar codes.
- FIG. 12 shows that one codeword includes only two information blocks by simplifying the example of FIG. 11 .
- the technical idea of FIG. 12 may be extended to a case where one codeword includes three or more information blocks.
- a receiver may decode a codeword and then perform a CRC check on each of the plurality of information blocks.
- CRC (1) is a CRC for information block 1 (info block 1)
- CRC (2) is a CRC for information block 2 (info block 2).
- a transmitter may retransmit only the first information block.
- the transmitter may retransmit only the second information block.
- Table 5 below shows retransmission methods in the success or failure of transmission when there are two CRCs respectively corresponding to the two information blocks shown in FIG. 12 .
- Table 5 summarizes available retransmission methods for each case of FIG. 12 . In this section, a case in which retransmission is allowed only one time will be described.
- the retransmission may be performed based on Q-learning, which will be described later.
- throughput may be defined as a performance index as shown in Equation 7.
- Equation 8 When the code rate is 0.5, if scheme 1 is used, the throughput and the upper limit thereof may be given by Equation 8.
- Equation 9 the throughput and the upper limit thereof may be given by Equation 9.
- Scheme 2 is effective when the SNR of a channel is low. This is because many errors occur in a poor channel environment.
- scheme 2 since retransmission is performed after polar coding, the probability of error detection may increase. However, in this case, since the entirety of a codeword is retransmitted, delay may also increase.
- scheme 1 is effective when the SNR of a channel is high. This is because errors are less likely to occur in a good channel environment. Thus, even if retransmission is performed without polar coding, it is expected that information bits are successfully decoded by combination of retransmitted information and initially transmitted information. In scheme 1, since the number of retransmitted bits is only half of the codeword (because the code rate is 0.5), transmission delay may decrease. In summary, when the channel environment is good, scheme 1 provides higher throughput.
- proposed scheme 1 (or scheme 3).
- the basic idea of this scheme is to retransmit only the first information block (without polar coding) upon retransmission in consideration of the fact that if a decoding error occurs in a previous information block during SIC decoding, the decoding error occurs also in a next information block.
- the throughput and the upper limit thereof may be given as follows.
- the transmitter may select a scheme with the best performance.
- the scheme with the best performance may vary not only depending on the statistical characteristics or channel gains of a channel but also depending on many parameters of a system.
- the retransmission problems may be optimized by the MAB algorithm.
- the value of Q for each possible action may be defined, managed, and updated to select the optimal action.
- Equation 10 shows how to update the value of Q in the MAB algorithm.
- Equation 10 the value of a denotes a step size or a learning rate and has a value between 0 and 1.
- R n denotes a reward.
- the value of ⁇ has the following tendency.
- ⁇ is too large (e.g., ⁇ 1), learning is not stable.
- the learning rate may vary depending on learning steps.
- the learning rate ⁇ may increase (may have a value close to 1) in the early stages of learning, and it may decrease (may have a value close to 0) in the later stages of the learning.
- the learning rate ⁇ may decrease monotonically as the learning progresses.
- the learning rate ⁇ may increase monotonically as the learning progresses.
- a opt ⁇ argmax a ⁇ A ⁇ Q ⁇ ( a ) , with ⁇ prob . 1 - ⁇ a ⁇ random ⁇ action , with ⁇ prob . ⁇ [ Equation ⁇ 11 ]
- Equation 11 may be referred to as a ⁇ -greedy algorithm in that for a probability of 1 ⁇ , an action a for maximizing the value of Q is selected based on the value of ⁇ according to the conventional greedy algorithm, and for a probability of ⁇ —, a random action is taken, unlike the general greedy algorithm.
- Equation 11 A denotes a set of all possible actions.
- the value of ⁇ is related to exploration and exploitation and has a value between 0 and 1. In this case, it is important to select and use the value of ⁇ . In general, the value of ⁇ has the following tendency.
- the exploration refers to a process of observing information about environments when there is no information.
- the exploitation refers to a process of applying learned results based on the observed information.
- the number of times that the exploitation is performed may decrease as the number of times that the exploration is performed increases.
- the number of times that the exploration is performed may increase, but the performance of the exploitation may be degraded.
- the number of times that the exploration is performed may be selected to maximize the expected reward.
- one of the following schemes may be performed: 1) transmitting information block 1 and information block 2 without applying polar coding (scheme 1); 2) transmitting information block 1 and information block 2 by applying polar coding (scheme 2); and 3) transmitting information block 1 and information block 2 by applying polar coding to information block 1 but without applying polar coding to information block 2 (scheme 3).
- the action set A may include scheme 1, scheme 2, and scheme 3 above.
- the reward is 0 when NACK occurs in any of the two information blocks.
- the value of the numerator is 1 when ACK occurs in the two information blocks.
- scheme 1 Since the value of the denominator of the reward is determined by delay, scheme 1 has the shortest delay, scheme 2 has the longest delay, and scheme 3 may have a value between scheme 1 and scheme 2.
- the value of Q may be determined by the reward value R and the previous value of Q.
- retransmission may be optimized based on the MAB algorithm.
- FIG. 13 is a diagram for explaining a NOMA system model with two users.
- FIG. 14 shows NOMA system method 1 (method 1) based on polar coding.
- each of Layer 1 and Layer 2 uses an independent polar code.
- FIG. 15 shows NOMA system method 2 (method 2) based on polar coding.
- two layers use one polar code.
- FIG. 16 is a diagram conceptually representing FIGS. 14 and 15 .
- FIG. 17 shows a retransmission method in a NOMA system using only one CRC for each layer. It is very difficult to determine which retransmission method among various possible retransmission methods through mathematical analysis. Based on the MAB algorithm, the optimal retransmission method may be determined.
- the reward function may be given as follows.
- the reward may be defined as follows.
- Equation 18 f(x 1 , x 2 , x 3 ) is an increasing function for x 1 , x 2 , and x 3 .
- the reward may be defined as follows based on linear combination.
- Equation 19 ⁇ 1 , ⁇ 2 , ⁇ 3 are positive constants.
- the optimal action may be determined by the greedy algorithm such as Equation 11.
- the value of ⁇ that determines the degree of exploration and exploration is an important parameter.
- many ways of changing the value of ⁇ over time have been proposed and studied.
- no method of actively changing the value of ⁇ depending on the characteristics of a radio channel has yet been proposed. Accordingly, the present disclosure proposes a method of actively changing the value of ⁇ depending on channel environments.
- FIG. 18 is a diagram for explaining a method of actively changing the value of depending on channel environments according to the present disclosure.
- the value of ⁇ may be configured to decrease as learning progresses.
- FIG. 18 shows such a concept, and more particularly, three cases in which the value of ⁇ value decreases.
- two possible equations are shown at the bottom of FIG. 18 . Since FIG. 18 is merely to show the concept, the value of ⁇ may be reduced by other equations or rules in real environments. The important point is that the value of ⁇ gradually decreases over time (that is, as the learning progresses).
- FIG. 19 is a diagram for explaining a method of actively changing the value of ⁇ depending on channel environments (e.g., channel coherence time) according to the present disclosure.
- the value of ⁇ may be configured to decrease at a high rate (i.e., high decaying speed). The reason for this is that when the channel changes quickly, the algorithm experiences various channel conditions within a relatively short time, so that rapid learning is possible. Thus, the value of ⁇ may be reduced more rapidly.
- the value of ⁇ may be configured to decrease at a row rate (i.e., low decaying speed). The reason for this is that when the channel changes slowly, the algorithm needs a lot of time to experience various channel states, so that learning is not fast. Thus, the value of ⁇ should be reduced more slowly.
- FIGS. 20 and 21 are diagrams for explaining a method of actively changing the value of depending on channel environments according to the present disclosure.
- the value of ⁇ may be reduced and maintained at the minimum value. However, when the statistical characteristics of the channel change, the transmitter may increase the value of ⁇ again.
- FIG. 20 shows a case in which the value of ⁇ increases when a change in the channel statistical characteristics is greater than a threshold
- FIG. 21 shows a case in which the value of ⁇ increases when the channel coherence time increases.
- the degree of decrease in the value of ⁇ is similar before and after state transition.
- the degree of decrease in the value of ⁇ also needs to change together.
- FIG. 22 is a diagram for explaining a method of actively changing the value of ⁇ depending on channel environments according to the present disclosure.
- the value of Q is updated according to Equation 10 as described above. In this case, if the value of ⁇ is too large, learning is unstable, and if the value of ⁇ is too small, learning is too slow.
- the value of ⁇ may be changed according to the following method.
- the learning rate ⁇ may have a relationship with a range of channel variation as shown in Equation 20.
- the learning rate ⁇ increases to enable fast learning.
- the value of ⁇ may increase because learning is generally much stable.
- the learning rate ⁇ decreases to enable stable learning.
- a HARQ procedure may be optimized by Q-learning.
- the HARQ procedure may be optimized on the assumption that channel size information
- Equation 22 may be satisfied.
- the HARQ procedure may start.
- j-th transmission i.e., (j ⁇ 1)-th retransmission
- u j coded bits included in a j-th subblock may be transmitted.
- throughput ⁇ may be defined as shown in Equation 23.
- T s is the length of each coded symbol (in seconds)
- NACK j denotes an event in which the receiver fails in decoding in all transmissions up to the j-th transmission (including the j-th transmission).
- the HARQ optimization relates to solving the following optimization problem.
- FIG. 23 is a diagram for explaining optimization of a HARQ procedure and a system model.
- FIG. 23 conceptually shows a HARQ system model and optimization thereof.
- Equation 23 a method of minimizing the number of retransmitted bits according to Equation 23 needs to be found, and the value of Pr(NACK j ) also needs to be calculated mathematically.
- the optimization problem is solved by assuming that the value of Pr(NACK j ) is represented by a relatively simple formula, does not change over time, and has the same formula for all users.
- Q-learning may be used.
- a set of states, a set of all possible actions, and a reward may be defined as shown in the following equations.
- the denominator of the reward represents delay due to retransmission (in the case of MAB), but the denominator of Equation 26 represents the total transmitted coded bits (in the case of Q-learning).
- the number of coded bits transmitted during the k-th transmission (or (k ⁇ 1)-th retransmission) may be used instead of delay.
- the overall Q-learning algorithm may be given as follows.
- the action A k is selected (for example, the ⁇ -greedy policy is used).
- a k ⁇ arg ⁇ ⁇ max ⁇ a ⁇ A ⁇ Q ⁇ ( S k , a ) , with ⁇ ⁇ prob . ⁇ 1 - ⁇ a ⁇ ⁇ random ⁇ ⁇ action , with ⁇ ⁇ prob . ⁇ ⁇ [ Equation ⁇ ⁇ 28 ]
- a k + 1 ⁇ arg ⁇ ⁇ max ⁇ a ⁇ A ⁇ Q ⁇ ( S k , a ) , with ⁇ ⁇ prob . ⁇ 1 - ⁇ a ⁇ ⁇ random ⁇ ⁇ action , with ⁇ ⁇ prob . ⁇ ⁇ [ Equation ⁇ ⁇ 31 ]
- Equation 32 a set of all actions A may be given by Equation 32.
- a all ⁇ a 1 ,a 2 ,a 3 ,a 4 ⁇ [Equation 32]
- a state may be defined as shown in Equation 33.
- S k ( k ,Case k ,N k ( a 1 ), N k ( a 2 ), N k ( a 3 ), N k ( a 4 )) [Equation 33]
- Equation 34 The set of all possible actions A that may be taken in the state S k may be defined by Equation 34.
- a reward R k+1 may be defined by Equation 35.
- the overall Q-learning algorithm may be given as follows.
- a k ⁇ arg ⁇ ⁇ max ⁇ a ⁇ A ⁇ Q ⁇ ( S k , a ) , with ⁇ ⁇ prob . ⁇ 1 - ⁇ a ⁇ ⁇ random ⁇ ⁇ action , with ⁇ ⁇ prob . ⁇ ⁇ [ Equation ⁇ ⁇ 36 ]
- An action A k+1 is selected as follows.
- a k + 1 ⁇ arg ⁇ ⁇ max ⁇ a ⁇ A ⁇ Q ⁇ ( S k , a ) , with ⁇ ⁇ prob . ⁇ 1 - ⁇ a ⁇ ⁇ random ⁇ ⁇ action , with ⁇ ⁇ prob . ⁇ ⁇ [ Equation ⁇ ⁇ 38 ]
- FIG. 24 shows a case in which one codeword includes four information blocks.
- a CRC may be added to each of the four information blocks, and polar coding-based retransmission may be performed as shown in Table 6 below.
- Table 6 shows retransmission methods depending on the success or failure of transmission of each information block when there are 4 CRCs respectively corresponding to the four information blocks.
- N k (a i ), i 1, 2, . . . ,
- : Number of times that action a i is selected and performed immediately before k-th transmission Set of actions A k : A k ⁇ possible retransmission in Case k ⁇ [Equation 40]
- FIG. 25 is a diagram for explaining a retransmission method in a NOMA system using one CRC in one layer.
- the coded bits of FIG. 25 may correspond to parity bits used in polar codes.
- information bits are equal to input bits of the polar coding, but related output bits of the polar coding may be different from the information bits.
- output bits related to information bits may be equal to input bits.
- FIG. 25 is a conceptual block diagram applicable to both non-systematic polar coding and systematic polar coding according to the characteristics of polar coding.
- the retransmission method may vary depending on the CRC check at each layer. Table 7 below shows various retransmission methods.
- Rx 1 may need to decode only Layer 1, but Rx 2 may need to decode both Layer 1 and Layer 2. If both the CRC (1) check and CRC (2) check are successful, no decoding is required.
- Rx 1 succeed in the CRC (1) check
- Rx 2 may fail in the CRC (1) check. Due to the characteristic of SIC, it is expected that Rx 2 will fail in the CRC (2) check upon failure of the CRC (1) check.
- Rx 2 may retransmit D (2) or the coded version of D (2) .
- D (2) when only D (1) is retransmitted, previously received D (2) may be used again for decoding according to the incremental redundancy method.
- FIG. 26 is a diagram for explaining a retransmission method in a NOMA system using a plurality of CRCs in one layer.
- the retransmission method may vary depending on CRC checks at each layer.
- the following table shows various retransmission methods in the NOMA system using a plurality of CRCs in one layer.
- a state, a set of actions, and a reward may be defined as shown in the following equations.
- R k + 1 ACK k ( 1 , 1 ) ⁇ [ 1 ] ⁇ ACK k ( 1 , 2 ) ⁇ [ 2 ] ⁇ ACK k ( 2 , 1 ) ⁇ [ 2 ] ⁇ ACK k ( 2 , 2 ) ⁇ [ 2 ] delay ⁇ ⁇ up ⁇ ⁇ to ⁇ ⁇ and ⁇ ⁇ including ⁇ ⁇ action ⁇ ⁇ A k [ Equation ⁇ ⁇ 47 ]
- retransmission performance may be improved by actively changing the value of ⁇ depending on channel environments as described above in sections 1.3.1, 1.3.2, and 1.3.3.
- retransmission performance may be improved by actively changing the value of ⁇ depending on channel environments as described above in section 1.3.4.
- DQN Several communication problems in HARQ retransmission based on polar coding and retransmission in NOMA may be solved by DQN.
- the reasons for using DQN instead of Q-learning are as follows.
- Q-learning the condition that the number of states needs to be finite should be satisfied.
- the number of states not only needs to be finite but also should not be too large to achieve satisfactory learning effects in Q-learning.
- the number of states may be very large or infinite. For example, it may be considered that a channel gain
- DQN may be applied to the aforementioned HARQ procedure using Q-learning.
- may be defined.
- a state, a set of actions, and a reward may be defined as shown in the following equations.
- the value of Q is not immediately updated, but may be indirectly updated by training an artificial neural network.
- w is a parameter of the artificial neural network
- the mean square error of these values may be defined as follows.
- J ⁇ ( w k ) E ⁇ [ ( R k + 1 + ⁇ ⁇ ⁇ max a ⁇ ⁇ Q ⁇ ( S k + 1 , a , w - ) - Q ⁇ ( S k , A k , w k ) ) 2 ] [ Equation ⁇ ⁇ 51 ]
- this value is copied from a learning artificial neural network at a regular interval.
- E[ ⁇ ] Past experience is stored in an experience replay memory. Then, the artificial neural network is trained by bring the past experience by a batch size according to a batch gradient method.
- the parameters of the artificial neural network may be updated as follows.
- w k+1 w k + ⁇ ′ ⁇ w k J ( w k ) [Equation 52]
- Equation 52 ⁇ ′ denotes the learning rate of the artificial neural network, and ⁇ denotes the gradient.
- the parameters of the neural network may be updated to maximize the expected reward value (see Equations 51 and 52).
- the parameters of the artificial neural network may mean parameters corresponding to lines connecting nodes at each layer of FIG. 9 .
- the computational complexity of DQN is greater than that of Q-learning, but it has advantages in that learning results are more accurate and a large amount of training data may be processed quickly in parallel.
- the value of ⁇ may be actively changed depending on channel environments to maximize performance as described above in sections 1.3.1, 1.3.2, and 1.3.3.
- DQN One of the most important features of DQN is the use of an experience replay memory. Specifically, a certain amount of experience may be randomly selected from the experience replay memory to update the artificial neural network according to a batch gradient method.
- the size of the experience replay memory may be adjusted adaptively or actively depending on the channel environments.
- the size of the experience replay memory may be reduced. If the channel characteristics frequently change, old experience that is no longer needed may be deleted from the memory because the old experience is not helpful for learning.
- the size of the experience replay memory should be kept large. The reason for this is that when the channel changes slowly, the correlation between experience samples is large. That is, since the large correlation needs to be reflected in learning, the size of the replay memory needs to be kept large.
- the size of the experience replay memory should be kept small In this way, when the statistical characteristics of the channel change, a new channel may be learned quickly.
- the second most important feature of DQN is that a target artificial neural network is separated from a learning artificial neural network, and the target artificial neural network is periodically updated.
- the update cycle of the target artificial neural network may be actively adjusted depending on channel environments.
- the update cycle of the target artificial neural network may be set sufficiently large to enable stable learning.
- the update cycle of the target artificial neural network may be shortened to enable fast learning. This may be interpreted to mean that the target artificial neural network, which has been trained based on the characteristics of an old channel, is discarded and a new target artificial neural network is used.
- the update cycle of the target artificial neural network may be lengthened to enable stable learning.
- the update cycle of the target artificial neural network need to be kept short. In this way, when the statistical characteristics of the channel change, a new channel may be learned quickly.
- Equation 53 the state may be defined as shown in Equation 53 to improve the performance of DQN.
- the state is defined as follows, channel gains when coded bits were transmitted may be included in the state, so that the transmitter may more accurately adjust its information transmission rate for each retransmission.
- State S k ( k ,( U 1 ,
- the discount factor ⁇ may be adjusted adaptively or dynamically.
- the discount factor is used in Equation 50.
- the discount factor ⁇ may be adjusted to reduce the impact of past data and increase the impact of current data as learning progresses. For example, when the statistical characteristics of a channel are fixed, the value of ⁇ may decrease at the beginning of learning, but may increase as time passes. When the statistical characteristics of the channel change, new learning is required, and thus the value of ⁇ may decrease and then gradually increase.
- the learning rate ⁇ ′ which is used for learning of the artificial neural network, may also be adjusted.
- the learning rate ⁇ ′ is to update the parameters of the artificial neural network. For example, when the statistical characteristics of a channel are fixed, the value of ⁇ ′ may decrease at the beginning of learning, but may increase as time passes. On the other hand, the value of ⁇ ′ may increase at the beginning of learning, but may decrease as time passes.
- the learning rate ⁇ ′ may be adjusted based on the statistical characteristics of the channel. When the statistical characteristics of the channel change, new learning is required, and thus the value of ⁇ ′ may decrease and then gradually increase. On the other hand, when the statistical characteristics of the channel change, the value of ⁇ ′ may increase and then gradually decrease.
- DQN may be applied when a pilot signal is transmitted as some of the coded bits of a polar code.
- FIG. 27 A is a diagram for explaining a method of combining a systematic polar code with channel measurement according to the present disclosure.
- FIG. 27 A is a diagram for explaining a method of increasing the efficiency of channel coding by combining channel estimation and systematic polar coding.
- a length-16 polar code is generated by combining a length-12 systematic polar code with four pilot signals is described.
- the four pilot signals may simultaneously serve two functions as follows.
- the actual code length becomes 16, and the coding rate is 8/16.
- the log likelihood ratio (LLR) values of received symbols corresponding to the four pilot signals are infinite, the actual code rate becomes 4/16, which is less than 1/3. Consequently, the reliability of information bits may be improved.
- pilot signals are transmitted at a regular interval. If the channel varies over time, transmitting pilot signals at a regular interval may be best in terms of channel estimation. However, such pilot signal allocation may not be best from the point of view of systematic polar codes.
- 4th, 8th, 12th, and 16th input signals (Rank 11, 5, 4, and 1) are converted into parity check bits (code bits) other than frozen bits. That is, 1st, 2nd, 3rd, 5th, 6th, 7th, 9th, and 10th input bits are used as frozen bits in systematic polar coding.
- the bit-channel capacity of the 10th input bit used as one frozen bit is 0.53274. Specifically, the 10th input bit corresponds to a bit channel with the 8th highest bit-channel capacity among the 16 input bits.
- the fundamental concept of polar codes is that input bits with high bit-channel capacities are used as information bits and input bits with low bit-channel capacities are used as frozen bits. That is, in the method of FIG. 27 A , the locations of the frozen bits (or the locations of the information bits) may not be optimized.
- FIG. 27 B is a diagram for explaining a method of combining a systematic polar code with channel measurement according to the present disclosure.
- a pilot signal may be transmitted as some coded bits in systematic polar coding.
- FIG. 27 B the problems mentioned in FIG. 27 A may be solved.
- FIG. 27 B is a diagram for explaining a method of combining a systematic polar code with channel measurement based on bit-channel capacities.
- frozen bits are transmitted on bit channels with the lowest bit-channel capacities among all input bits.
- pilot signals may not be arranged at a regular interval. Permutation operation may be applied to solve this problem, that is, to arrange pilot signals at a regular interval.
- the LLR values thereof may become infinite during decoding.
- x (x P , x A,F ): polar coding output bit vector (codeword)
- Equation 54 is established from the polar code generator matrix.
- Decoding of systematic polar codes is basically equal to decoding of non-systematic polar codes. However, since the receiver knows codeword symbols corresponding to pilot signals, the LLR values thereof may be set infinite.
- FIG. 28 A is a diagram for explaining a method of combining a non-systematic polar code with channel measurement according to the present disclosure.
- uF, and uA correspond to xAF, and up corresponds to xP as described above.
- the correspondence means that the locations of input bits (uF, uA, and uP) are related to those of output bits (xAF and xP).
- input bits are converted into output bits based on the polar code generator matrix.
- the bit values of input bits set as frozen bits may be 0, and pilot signal bits may be bit sequences used as pilot signals. That is, coded bits may be determined such that the pilot signal bits are equivalent to the bit sequences used as pilot signals.
- pilot signal bits among output bits obtained from polar coding may be known bits.
- FIG. 28 B is a diagram for explaining a method of combining a non-systematic polar code with channel measurement where permutation is used to arrange pilot signals at a regular interval according to the present disclosure.
- a pilot signal may be transmitted as some coded bits in non-systematic polar coding.
- pilot insertion may be used for both systematic and non-systematic polar codes in polar coding.
- FIG. 28 B shows a method of arranging pilot signals at a regular interval by additionally applying permutation operation to the method of FIG. 28 A .
- the locations of coded bits for generating pilot signal bits are determined by channel capacity.
- the pilot signal bits are generated such that the locations thereof correspond to those of the coded bits, the distance between the locations of the pilot signal bits among polar coding output bits is determined by the channel capacity.
- the locations of the pilot signal bits may be changed by the permutation operation.
- the most important question is how many pilot signals need to be included to obtain the best performance when the length of a codeword is given. If too few pilot signals are inserted, the overall decoding performance may be degraded due to channel estimation errors. On the contrary, if pilot signals are inserted more than necessary, the number of coded bits that we actually send decreases even though channel estimation becomes accurate, so that the overall performance may be degraded as well. Therefore, it is important to find the optimal number of pilot signals inserted in the polar code. However, it is very difficult to solve this problem with an analytical method based on mathematics or a dynamic programming method. This is because coding and all system parameters affect the performance In polar coding, the number of pilots to be inserted may be calculated based on DQN.
- a state, a set of actions, and a reward may be defined as follows for DQN configured to calculate the number of inserted pilots.
- Set of actions A: A ⁇ 1,2, . . . , ⁇ pilot max ⁇ [Equation 57]
- Equation 57 ⁇ pilot max denotes the maximum number of pilot signals inserted into one codeword.
- R k+1 ( n ⁇ A k ⁇ l p ) ⁇ ACK k [Equation 58]
- DQN may be applied to the retransmission and HARQ in polar codes described above in section 2.2 above. As shown in FIG. 12 , it is assumed that there are two information blocks.
- a state and a set of all actions in DQN may be defined as follows.
- Set of all possible actions, A all ⁇ a 1 ,a 2 ,a 3 ,a 4 ⁇ [Equation 59]
- DQN may be applied by defining the state and the action set in order to solve the aforementioned problems.
- DQN may be applied to the above-described NOMA HARQ.
- a state may be defined for DQN.
- State S k ( k ,Case k ,N k ( a 1 ), N k ( a 2 ), N k ( a 3 ), N k ( a 4 ), N k ( a 5 ), N k ( a 6 ),
- ) or S k ( k ,Case k ,N k ( a 1 ), N k ( a 2 ), . . . , N k ( a 6 ),
- N k (a i ), i 1, 2, 3, 4, 5, 6: Number of times that action a i is selected and performed immediately before k-th transmission
- DQN may be applied by defining the state and the action set as described above.
- FIG. 29 is a diagram for explaining a method of transmitting data based on polar coding according to the present disclosure.
- the present disclosure provides a method for transmitting data based on polar coding in a wireless communication system.
- the method may include: transmitting data including a plurality of information blocks, wherein each of the plurality of information blocks includes a corresponding CRC; receiving a hybrid automatic repeat request acknowledgement/negative acknowledgment (HARQ ACK/NACK) for the transmitted data; learning to retransmit the plurality of information blocks; and retransmitting the plurality of information blocks based on the HARQ ACK/NACK.
- the learning may include: obtaining a current state s n ; obtaining actions to be applied to the current state s n ; and selecting an action for maximizing an expected reward value Q n+1 from among the actions.
- the expected reward value Q n+1 may be obtained based on rewards R 1 , R 2 , . . . , R n respectively related to states s 1 , s 2 , . . . , s n .
- the plurality of information blocks may be retransmitted based on the selected action.
- the learning may further include obtaining a next state s n+1 based on the current state s n and the selected action.
- the learning step may be repeatedly performed.
- next state s 2 may be obtained by performing the learning based on the current state s 1 , and the learning may be repeatedly performed so that the states s 1 , s 2 , . . . , s n may be obtained.
- a reward corresponding to the current state may be obtained based on an action selected in the current state. Considering that a plurality of actions may be applied to the current state, in actual learning, all rewards for all actions may be obtained, and then based on rewards obtained so far, a reward (and action related thereto) for maximizing the expected reward value may be selected from among all the rewards. The expected reward value may be obtained based on the rewards obtained so far and the reward related to the selected action.
- the current state may include information on the number of times that transmission is performed (e.g., k-th transmission) and information on the number of bits transmitted up to now.
- the expected reward value Q n+1 may be defined by the following equation based on a latest reward R n among the rewards R 1 , R 2 , . . . , R n and a previous expected reward value Q n .
- Q n+1 ⁇ n R n +(1 ⁇ ) Q n
- a learning rate ⁇ may be determined based on a range of channel variation.
- the actions may include a first action of transmitting the plurality of information blocks without coding, a second action of coding and transmitting the plurality of information blocks, and a third action of coding and transmitting some of the plurality of information blocks and transmitting the remaining blocks without coding.
- Each of the rewards related to the states may be obtained based on the cumulative number of bits of multiple information blocks transmitted up to now and the HARQ ACK/NACK.
- the cumulative number of bits of multiple information blocks transmitted up to now and the HARQ ACK/NACK may be obtained based on the current state and the selected action.
- the expected reward value Q n+1 may be a weighted average of the rewards based on a learning rate, and the learning rate may decrease monotonically as the learning progresses.
- the expected reward value Q n+1 may be defined by the following equation based on the rewards R 1 , R 2 , . . . , R n .
- the expected reward value Q n+1 may be defined by the following equation based on a latest reward R n among the rewards R 1 , R 2 , . . . , R n and a previous expected reward value Q n .
- Q n+1 ⁇ n R n +(1 ⁇ ) Q n
- a learning rate ⁇ n may decrease monotonically as n increases.
- the learning rate ⁇ n may increase monotonically as n increases.
- the expected reward value may be referred to as ‘Q value’ in Q-learning or DQN.
- a MAB algorithm For the learning in the method for processing retransmission based on polar coding, a MAB algorithm, a Q-learning algorithm, and a DQN algorithm may be used.
- these algorithms may be used for retransmission in NOMA communication systems.
- the actions may include a first action of transmitting the plurality of information blocks without coding, a second action of coding and transmitting the plurality of information blocks, a third action of coding and transmitting some of the plurality of information blocks, and a fourth action of transmitting some of the plurality of information blocks without coding.
- a states may include information on the number of times that retransmission is performed (k), information about a situation that an agent encounters (for example, cases in Table 6 or 7), and information about the number of times that each action has been performed up to now.
- the present disclosure provides an apparatus for transmitting data based on polar coding in a wireless communication system.
- the apparatus may include: a transceiver; a memory; and at least one processor connected to the transceiver and the memory.
- the memory may be configured to store instructions that, when executed, cause the at least one processor to perform operations including: transmitting data including a plurality of information blocks, wherein each of the plurality of information blocks includes a corresponding CRC; receiving a HARQ ACK/NACK for the transmitted data; learning to retransmit the plurality of information blocks; and retransmitting the plurality of information blocks based on the HARQ ACK/NACK.
- the learning may include: obtaining a current state s n ; obtaining actions to be applied to the current state s n ; and selecting an action for maximizing an expected reward value Q n+1 from among the actions.
- the expected reward value Q n+1 may be obtained based on rewards R 1 , R 2 , . . . , Rn respectively corresponding to states s 1 , s 2 , . . . , s n .
- the plurality of information blocks may be retransmitted based on the selected action.
- the apparatus may be mounted on an autonomous driving apparatus configured to communicate with at least one of a mobile terminal, a base station, or an autonomous driving vehicle.
- the present disclosure provides a method for transmitting data based on polar coding in a wireless communication system.
- the method may include: transmitting data including a plurality of information blocks, wherein each of the plurality of information blocks includes a corresponding CRC; receiving a HARQ ACK/NACK for the transmitted data; and retransmitting the plurality of information blocks based on the HARQ ACK/NACK.
- the method may further include learning to retransmit the plurality of information blocks.
- the learning may use at least one of a MAB algorithm, a Q-learning algorithm, or a DQN algorithm.
- the learning may include: obtaining a current state s n ; obtaining actions to be applied to the current state s n ; and selecting an action for maximizing an expected reward value Q n+1 from among the actions.
- the expected reward value Q n+1 may be obtained based on rewards R 1 , R 2 , . . . , R n respectively related to states s 1 , s 2 , . . . , s n .
- the plurality of information blocks may be retransmitted based on the selected action.
- the learning may further include obtaining a next state s n+1 based on the current state s n and the selected action.
- the present disclosure provides a method for transmitting data based on polar coding in a wireless communication system.
- the method may include: obtaining the number of pilot bits transmitted with a data sequence; generating encoded bits by encoding the data sequence and the pilot bits based on a polar code; and transmitting the encoded bits.
- Obtaining of the number of pilot bits transmitted with the data sequence may further include learning based on at least one of a MAB algorithm, a Q-learning algorithm, or a DQN algorithm.
- the learning may include: obtaining a current state s n ; obtaining actions to be applied to the current state s n ; and selecting an action for maximizing an expected reward value Q n+1 from among the actions.
- the expected reward value Q n+1 may be obtained based on rewards R 1 , R 2 , . . . , R n respectively corresponding to states s 1 , s 2 , . . . , s n .
- the learning may further include obtaining a next state s n+1 based on the current state s n and the selected action.
- FIG. 30 is a diagram for explaining a method of receiving data based on polar coding according to the present disclosure.
- the present disclosure provides a method for receiving data based on polar coding in a wireless communication system.
- the method may include: receiving data including a plurality of information blocks, wherein each of the plurality of information blocks includes a corresponding CRC; transmitting a HARQ ACK/NACK for the received data; learning to receive again the plurality of information blocks; and receiving again the plurality of information blocks based on the HARQ ACK/NACK.
- the learning may include: obtaining a current state s n ; obtaining actions to be applied to the current state s n ; and selecting an action for maximizing an expected reward value Q n+1 from among the actions.
- the expected reward value Q n+1 may be obtained based on rewards R 1 , R 2 , . . .
- the plurality of information blocks may be retransmitted based on the selected action.
- the plurality of information blocks may be received again based on the selected action.
- the present disclosure provides an apparatus for processing retransmission based on polar coding in a wireless communication system.
- the apparatus may include: at least one processor; and a transceiver connected to the at least one processor.
- the at least one processor may be configured to: receive data including a plurality of information blocks, wherein each of the plurality of information blocks includes a corresponding CRC; transmit a HARQ ACK/NACK for the received data; learn to receive again the plurality of information blocks; and receive again the plurality of information blocks based on the HARQ ACK/NACK.
- the learning may include: obtaining a current state s n ; obtaining actions to be applied to the current state s n ; and selecting an action for maximizing an expected reward value Q n+1 from among the actions.
- the expected reward value Q n+1 may be obtained based on rewards R 1 , R 2 , . . . , R n respectively corresponding to states s 1 , s 2 , . . . , s n .
- the plurality of information blocks may be retransmitted based on the selected action.
- the plurality of information blocks may be received again based on the selected action.
- the method and apparatus for performing channel coding based on polar codes are industrially applicable to various wireless communication systems including 3GPP LTE/LTE-A systems, 5G communication systems, etc.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Theoretical Computer Science (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
Description
Q n+1 =αR n+(1−α)Q n
Q n+1=αn R n+(1−α)Q n
TABLE 1 | ||||
μ | Δf = 2μ*15 [kHz] | Cyclic prefix(CP) | ||
0 | 15 | |
||
1 | 30 | |
||
2 | 60 | Normal, Extended | ||
3 | 120 | |
||
4 | 240 | Normal | ||
TABLE 2 | ||||
Frequency Range | Corresponding | Subcarrier Spacing | ||
designation | frequency range | |||
FR1 | 450MHz - |
15, 30, 60kHz | ||
FR2 | 24250MHz - |
60, 120, 240kHz | ||
TABLE 3 | |||||
μ | Nslot symb | Nframe,μ slot | Nsubframe,μ slot | ||
0 | 14 | 10 | 1 | ||
1 | 14 | 20 | 2 | ||
2 | 14 | 40 | 4 | ||
3 | 14 | 80 | 8 | ||
4 | 14 | 160 | 16 | ||
TABLE 4 | |||||
μ | Nslot symb | Nframe,μ slot | Nsubframe,μ slot | ||
2 | 12 | 40 | 4 | ||
-
- For a PCell downlink, offsetToPointA represents the frequency offset between point A and the lowest subcarrier of the lowest resource block overlapping with an SS/PBCH block used by the UE for initial cell selection, which is expressed in units of resource blocks on the assumption of an SCS of 15 kHz for FR1 and an SCS of 60 kHz for FR2.
- For other cases, absoluteFrequencyPointA represents the frequency location of point A expressed as in the absolute radio-frequency channel number (ARFCN).
n CRB =n PRB +N BWP,i start [Equation 2]
- E. Arikan, “Channel polarization: A method for constructing capacity achieving codes for symmetric binary-input memoryless channels,” IEEE Transaction on Information Theory, vol. 55, no. 7, pp. 3051-3073, July 2009.
- E. Arikan, “Systematic polar coding,” IEEE Communication Letters, vol. 15, pp. 860-862, August 2011.
TABLE 5 | ||||
Case | Erroneous blocks | Possible Retransmissions (actions) | ||
1 | O, O | No retransmission | ||
2 | X, O | {1} | ||
3 | O, X | {2} | ||
4 | X, X | {1, 2}, coded {1, 2} | ||
A={
Q(a)=Q(a)+α(R−Q(a))=(1−a)Q(a)+αR [Equation 14]
α∈A={(D (1)),{D (1) ,D (2)},coded{D (1)},coded{D (1) ,D (2)}} [Equation 15]
Q(a)=Q(a)+α(R−Q(a))=(1−α)Q(a)+αR [Equation 16]
State S k:
S k=(k,U k−1) [Equation 25]
-
- k∈{1, 2, . . . , J}
- J: Maximum number of allowed retransmissions
- Uk=Σj=1 kuj: Sum of numbers of all coded bits transmitted up to k-th transmission
Set of all actions, A:
A={b 1 ,b 2 , . . . ,b L} [Equation 26]
-
- If decoding succeeds due to an action Ak, ACKk=1.
- If decoding fails due to the action Ak, ACKk=0.
A all ={a 1 ,a 2 ,a 3 ,a 4} [Equation 32]
-
- a1: Retransmission of {1}
- a2: Retransmission of {2}
- a3: Retransmission of {1,2}
- a4: Retransmission of coded {1,2}
S k=(k,Casek ,N k(a 1),N k(a 2),N k(a 3),N k(a 4)) [Equation 33]
-
- Where, k∈{1, 2, . . . , J}
- Casek∈{2, 3, 4}; Case corresponding to time when k-th transmission is performed
- Nk(ai), i=1, 2, 3, 4: Number of times that action αi is selected and performed immediately before k-th transmission
TABLE 6 | ||
Case | Erroneous blocks | Possible Retransmissions (actions) |
1 | O, O, O, O | No retransmission |
2 | X, O, O, O | {1} |
3 | O, X, O, O | {2} |
4 | O, O, X, O | {3} |
5 | O, O, O, X | {4} |
6 | X, X, O, O | {1}, {1, 2}, their coded versions |
7 | X, O, X, O | {1}, {1, 3}, their coded versions |
8 | X, O, O, X | {1}, {1, 4}, their coded versions |
9 | O, X, X, O | {2}, {2, 3}, their coded versions |
10 | O, X, O, X | {2}, {2, 4}, their coded versions |
11 | O, O, X, X | {3}, {3, 4}, their coded versions |
12 | X, X, X, O | {1}, {1, 2}, {1, 2, 3}, {1, 3} their coded versions |
13 | X, X, O, X | {1}, {1, 2}, {1, 2, 4}, {1, 4} their coded versions |
14 | X, O, X, X | {1}, {1, 3}, {2, 3, 4}, {2, 4} their coded versions |
15 | O, X, X, X | {2}, {2, 3}, {2, 3, 4}, {2, 4} their coded versions |
16 | X, X, X, X | {1}, {1, 2}, {1, 2, 3}, {1, 3}, {1, 3, 4}, |
{1, 2, 4}, {1, 2, 3, 4}, their coded versions | ||
State S k:
S k=(k,Casek ,N k(a 1),N k(a 2), . . . ,N k(a |A
Set of actions A k:
A k={possible retransmission in Casek} [Equation 40]
TABLE 7 | ||||
| Rx | 1 | |
|
1 | CRC(1) pass | CRC(2) pass | No |
|
2 | CRC(1) pass | CRC(1) pass | {D(2)}, or its coded version | |
CRC(2) fail | ||||
3 | CRC(1) pass | CRC(1) fail | {D(1)}, { D(1), D(2)}, or its coded version | |
CRC(2) fail | ||||
4 | CRC(1) fail | CRC(1) pass | {D(1)}, { D(1), D(2)}, or its coded version | |
CRC(2) fail | ||||
5 | CRC(1) fail | CRC(1) fail | {D(1)}, { D(1), D(2)}, or its coded version | |
CRC(2) fail | ||||
State S k:
S k=(k,Casek ,N k(a 1),N k(a 2), . . . ,N k(a 6)) [Equation 42]
Set of actions A k:
A k={possible retransmission in Casek} [Equation 43]
-
- If the CRC(i) check succeeds by tacking the action Ak for user m, ACKk (i)[m]=1.
- If the CRC(i) check fails by tacking the action Ak for user m, ACKk (i)[m]=0.
TABLE 8 | ||||
| Rx | 1 | |
|
1 | CRC(1,1) pass | CRC(2,1) pass | No retransmission | |
CRC(1,2) pass | CRC(2,1) pass | |||
2 | CRC(1,1) pass | CRC(2,1) pass | {D(2,2)}, or its coded version | |
CRC(1,2) pass | CRC(2,1) fail | |||
... | ... | ... | ... | |
M | CRC(1,1) fail | CRC(1,1) fail | {D(1,1)}, { D(1,1), D(1,2), {D(1,1), D(1,2), | |
CRC(1,2) fail | CRC(1,2) fail | D(2,1)} | ||
CRC(2,1) fail | {D(1,1), D(1,2), D(2,1), D(2,2)}, | |||
CRC(2,1) fail | or its coded version | |||
State S k:
S k=(k,Casek ,N k(a 1),N k(a 2), . . . ,N k(a |A
Set of actions A k:
A k={possible retransmission in Casek} [Equation 46]
-
- If the CRC(i) check succeeds by tacking the action Ak for user m, ACKk (i)[m]=1.
- If the CRC(i) check fails by tacking the action Ak for user m, ACKk (i)[m]=0.
State S k:
S k=(k,U k−1 ,|h k|) [Equation 48]
-
- k∈{1, 2, . . . , J}
- Uk=Σj=1 k uj: Sum of numbers of all coded bits transmitted up to k-th transmission
Set of all actions A:
A={b 1 ,b 2 , . . . ,b L} [Equation 49]
-
- If decoding succeeds due to an action Ak, ACKk=1.
- If decoding fails due to action Ak, ACKk=0.
w k+1 =w k+α′∇w
State S k:
S k=(k,(U 1 ,|h 1|),(U 2 ,|h 2|), . . . ,(U k−1 ,|h k−1|),|h k|) [Equation 53]
-
- A: Index set of information bits
- P: Index set of pilot signal bits
- F: Index set of frozen bits
-
- xP: This denotes a vector set of pilot signal bits in a codeword corresponding to bit indices, that is, an output bit vector in a codeword positioned at the same location as an input bit vector up. In other words, the input bit vector up is related to the output bit vector XP.
- xA,F: This denotes an output bit vector in a codeword positioned at the same location of input bit vectors uA and uF. That is, the output bit vector xA,F is related to input bit vectors uA and up.
-
- uA: This denotes an input bit vector related to the indices of information bits.
- uP: This denotes a vector of input bits positioned at the same location as pilot signals in a codeword related to the indices of pilot signal bits.
- uF: This denotes a frozen bit vector related to the indices of frozen bits.
-
- G: This denotes a polar code (coding) generator matrix.
- GAF,P: This denotes a sub-matrix of G, where an (i, j)-th element Gi,j may satisfy the following condition: i∈A∪F and j∈P.
- GP,P, GP,AF, and GAF,AF: These are defined similarly to GAF,P.
x P =u P G P,P+(u A ,u F)G AF,P
x AF =u P G P,AF+(u A ,u F)G AF,AF
u P=(x P−(u A ,u F)G AF,P)G P,P −1 (step 1):
x AF =u P G P,AF+(u A ,u F)G AF,AF (step 2):
State S k:
S k=(k,|h k|) [Equation 56]
Set of actions A:
A={1,2, . . . ,Ωpilot max} [Equation 57]
Reward R k+1:
R k+1=(n−A k ×l p)×ACK k [Equation 58]
-
- If decoding succeeds, ACKk=1.
- If decoding fails, ACKk=0.
- n: Codeword length (number of coded bits in codeword)
- Length of each pilot (number of bits used for each pilot)
Set of all possible actions,A all:
A all ={a 1 ,a 2 ,a 3 ,a 4} [Equation 59]
-
- a1: Retransmission of {1}
- a2: Retransmission of {2}
- a3: Retransmission of {1,2}
- a4: Retransmission of coded {1,2}
State S k:
S k=(k,Casek ,N k(a 1),N k(a 2)N k(a 3)N k(a 4),|h k|)
or
S k=(k,Casek ,N k(a 1),N k(a 2)N k(a 3)N k(a 4),|h k−L |,|h k−L+1 |, . . . ,|h k|) [Equation 60] - k∈{1, 2, . . . , J}
- Casek∈{2, 3, 4}; Case corresponding to time when k-th transmission is performed
- Nk(ai), i=1, 2, 3, 4: Number of times that action ai is selected and performed immediately before k-th transmission
State S k:
S k=(k,Casek ,N k(a 1),N k(a 2),N k(a 3),N k(a 4),N k(a 5),N k(a 6),|h k|)
or
S k=(k,Casek ,N k(a 1),N k(a 2), . . . ,N k(a 6),|h k−L |,|h k−L+1 |, . . . ,|h k|) [Equation 61]
-
- k∈{1, 2, . . . , J}
Q n+1=αn R n+(1−α)Q n
Q n+1=αn R n+(1−α)Q n
Claims (16)
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR20180155495 | 2018-12-05 | ||
KR10-2018-0155495 | 2018-12-05 | ||
PCT/KR2019/017092 WO2020116958A1 (en) | 2018-12-05 | 2019-12-05 | Method and apparatus for transmitting data on basis of polar coding in wireless communication system |
Publications (2)
Publication Number | Publication Date |
---|---|
US20220029638A1 US20220029638A1 (en) | 2022-01-27 |
US12003254B2 true US12003254B2 (en) | 2024-06-04 |
Family
ID=70974778
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US17/297,705 Active 2041-03-29 US12003254B2 (en) | 2018-12-05 | 2019-12-05 | Method and apparatus for transmitting data on basis of polar coding in wireless communication system |
Country Status (2)
Country | Link |
---|---|
US (1) | US12003254B2 (en) |
WO (1) | WO2020116958A1 (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2021150919A (en) * | 2020-03-23 | 2021-09-27 | ソニーグループ株式会社 | Communication device and communication method |
US11562174B2 (en) * | 2020-05-15 | 2023-01-24 | Microsoft Technology Licensing, Llc | Multi-fidelity simulated data for machine learning |
CN113810155B (en) * | 2020-06-17 | 2022-11-18 | 华为技术有限公司 | Channel coding and decoding method and communication device |
CN112616158B (en) * | 2020-12-14 | 2023-09-05 | 中国人民解放军空军工程大学 | Cognitive communication interference decision method |
WO2024103298A1 (en) * | 2022-11-16 | 2024-05-23 | 华为技术有限公司 | Data transmission method and apparatus |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20160115803A (en) | 2015-03-25 | 2016-10-06 | 삼성전자주식회사 | Apparatus and method for constructing harq rate-compatible polar codes |
KR20170086640A (en) | 2014-11-27 | 2017-07-26 | 후아웨이 테크놀러지 컴퍼니 리미티드 | Rate matching method and apparatus for polar code, and wireless communications device |
US20190019082A1 (en) * | 2017-07-12 | 2019-01-17 | International Business Machines Corporation | Cooperative neural network reinforcement learning |
US20190052418A1 (en) * | 2016-03-18 | 2019-02-14 | Qualcomm Incorporated | Transmission of new data in a hybrid automatic repeat request (harq) retransmission with polar coded transmissions |
US20210258988A1 (en) * | 2018-09-28 | 2021-08-19 | Intel Corporation | System and method using collaborative learning of interference environment and network topology for autonomous spectrum sharing |
-
2019
- 2019-12-05 US US17/297,705 patent/US12003254B2/en active Active
- 2019-12-05 WO PCT/KR2019/017092 patent/WO2020116958A1/en active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20170086640A (en) | 2014-11-27 | 2017-07-26 | 후아웨이 테크놀러지 컴퍼니 리미티드 | Rate matching method and apparatus for polar code, and wireless communications device |
KR20160115803A (en) | 2015-03-25 | 2016-10-06 | 삼성전자주식회사 | Apparatus and method for constructing harq rate-compatible polar codes |
US20190052418A1 (en) * | 2016-03-18 | 2019-02-14 | Qualcomm Incorporated | Transmission of new data in a hybrid automatic repeat request (harq) retransmission with polar coded transmissions |
US20190019082A1 (en) * | 2017-07-12 | 2019-01-17 | International Business Machines Corporation | Cooperative neural network reinforcement learning |
US20210258988A1 (en) * | 2018-09-28 | 2021-08-19 | Intel Corporation | System and method using collaborative learning of interference environment and network topology for autonomous spectrum sharing |
Non-Patent Citations (3)
Title |
---|
Peihong Yuan, "Flexible IR-HARQ scheme for polar-coded modulation", IEEE Wireless Communications and Networking Conference Workshops (WCNCW), IEEE, 2018, p. 49-54, May 31, 2018. |
Shao-Yu Lien, "Optimum Ultra-Reliable and Low Latency Communications in 5G New Radio", Mobile Networks and Applications, Springer, 2018, vol. 23, p. 1020-1027, Nov. 2, 2017. |
Wen-Ching Chung, "HARQ Control Scheme by Fuzzy Q-Learning for HSPA+", IEEE 73rd Vehicular Technology Conference (VTC Spring), IEEE, Jul. 18, 2011. |
Also Published As
Publication number | Publication date |
---|---|
US20220029638A1 (en) | 2022-01-27 |
WO2020116958A1 (en) | 2020-06-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US12003254B2 (en) | Method and apparatus for transmitting data on basis of polar coding in wireless communication system | |
US11239947B2 (en) | Bit allocation for encoding and decoding | |
US10461779B2 (en) | Rate-compatible polar codes | |
CN110249558B (en) | Polarization code for downlink control channel of wireless network | |
US10833706B2 (en) | Polar encoding and decoding using predefined information | |
US20230090593A1 (en) | Method and device for transmitting and receiving wireless signal in wireless communication system | |
US20180331697A1 (en) | Nominal complexity and weighted combinations for polar code construction | |
US11956077B2 (en) | Method and device for performing channel coding on basis of polar coding in wireless communication system | |
CN107124188A (en) | Coding method, interpretation method, encoding device and the decoding equipment of polarization code | |
KR20170074684A (en) | Method and apparatus for encoding in wireless communication system | |
US20210013902A1 (en) | Device and method to transmit and receive signal in communication system | |
CN111164914B (en) | Hybrid decoder for slotted ALOHA coding | |
CN109792314A (en) | The retransmission technique of transmission for coding | |
CN110402553A (en) | Coding techniques for Multistage Control information | |
US10819472B2 (en) | Method for performing HARQ by using polar code having random length | |
US11616598B2 (en) | Puncturing and retransmission techniques for encoded transmissions | |
US11595155B2 (en) | Method for decoding low density parity check (LDPC)-coded signal, and terminal therefor | |
CN110622426B (en) | Puncturing polar codes with complementary sequences | |
US11963023B2 (en) | Early negative acknowledgment indication for a data message | |
US20230082053A1 (en) | Method and apparatus for transceiving and receiving wireless signal in wireless communication system | |
US11784663B2 (en) | Method and apparatus for performing encoding on basis of parity check matrix of low density parity check code generated from protograph in wireless communication system | |
CN111490798B (en) | Decoding method and decoding device | |
US11675658B2 (en) | Method and apparatus for decoding signal in wireless communication system | |
WO2021159440A1 (en) | Techniques for determining a degree distribution in a multi-hop network | |
US10778252B2 (en) | Retransmission scheme for low-density parity check coding |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: LG ELECTRONICS INC., KOREA, REPUBLIC OF Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:KIM, BONGHOE;NOH, KWANGSEOK;KIM, ILMIN;SIGNING DATES FROM 20210329 TO 20210402;REEL/FRAME:056408/0851 |
|
FEPP | Fee payment procedure |
Free format text: ENTITY STATUS SET TO UNDISCOUNTED (ORIGINAL EVENT CODE: BIG.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: DOCKETED NEW CASE - READY FOR EXAMINATION |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: NON FINAL ACTION MAILED |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: NOTICE OF ALLOWANCE MAILED -- APPLICATION RECEIVED IN OFFICE OF PUBLICATIONS |
|
ZAAB | Notice of allowance mailed |
Free format text: ORIGINAL CODE: MN/=. |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: NOTICE OF ALLOWANCE MAILED -- APPLICATION RECEIVED IN OFFICE OF PUBLICATIONS |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: PUBLICATIONS -- ISSUE FEE PAYMENT RECEIVED |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: AWAITING TC RESP, ISSUE FEE PAYMENT RECEIVED |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: PUBLICATIONS -- ISSUE FEE PAYMENT RECEIVED |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: PUBLICATIONS -- ISSUE FEE PAYMENT VERIFIED |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |